Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Minimally Invasive Bipolar Fractional Radiofrequency Treatment Upregulates Anti-Senescence Pathways
Article first published online: 26 APR 2013
Copyright © 2013 Wiley Periodicals, Inc.
Lasers in Surgery and Medicine
Volume 45, Issue 4, pages 201–206, April 2013
How to Cite
Rangarajan, S., Trivedi, A., Ubeid, A. A. and Hantash, B. M. (2013), Minimally Invasive Bipolar Fractional Radiofrequency Treatment Upregulates Anti-Senescence Pathways. Lasers Surg. Med., 45: 201–206. doi: 10.1002/lsm.22135
- Issue published online: 26 APR 2013
- Article first published online: 26 APR 2013
- Manuscript Accepted: 13 MAR 2013
- bipolar fractional radiofrequency;
The sirtuin gene family has been implicated in various anti-senescence pathways. Its connection, if any, with the skin wound healing response has yet to be elucidated.
The goal of our study was to better understand the effects of FRF treatment on the sirtuin anti-senescence pathway in skin.
Human abdominal skin was treated with FRF, and then harvested at 0, 2, 14, and 28 days post-treatment to assess for temporal changes in gene expression levels.
Decreased levels of SIRT1, 3, 5, and 7 were observed immediately post-FRF treatment. By Day 2, SIRT1, 6, and 7 expressions increased 50–100%. SIRT6 and 7 expression continued to increase through Day 28. Expression levels of apoptosis genes FoxO3 and p53 decreased, while Bax levels increased by Day 28.
Our results raise the possibility that sirtuin activity may be used as an accurate corollary to clinical improvement in skin quality. Lasers Surg. Med. 9999:XX–XX, 2013. © 2013 Wiley Periodicals, Inc.
interfollicular epidermal stem cells
Skin aging results from a combination of extrinsic (e.g., chemical, toxins, pollutants, ultraviolet light, and ionizing radiation) and intrinsic factors (e.g., gene mutations, cellular metabolism, and hormone environment) . These factors manifest clinically as wrinkles, uneven pigmentation and loss of firmness. Histologically, epidermal and dermal atrophy are observed.
UV damage accelerates epidermal stem cell senescence and decreases the number and synthetic capacity of fibroblasts, the dermal cell responsible for production of extracellular matrix components . UV damage also upregulating AP-1, which blocks transforming growth factor-β-mediated collagen gene transcription . Thus, the harmful effects of UV radiation have led to the development of a wide array of treatments aimed at reversing photodamage. These include topical retinoids, chemical peels, microdermabrasion, and the use of noninvasive electromagnetic radiation (EMR) devices (e.g., lasers, light emitting diodes, intense pulsed light, and ultrasound) . More recently, we examined the efficacy of a novel, minimally invasive, microneedle-based bipolar fractional radiofrequency (FRF) system for treatment of facial skin laxity and rhytides . The mechanisms underlying this reversal are thought to involve initiation of dermal remodeling leading to removal of photoaged tissue and its replacement with new collagen and elastin [6-8].
However, studies have indicated that there is not always a direct correlation between collagen denaturation and/or levels and the extent of clinical improvement in skin quality . Furthermore, recent clinical data also showed that FRF treatment also resulted in overall skin textural improvement , suggesting that it also induced anti-aging effects on the epidermal layer.
The above findings have led some investigators to seek novel biomarkers that better correlate with clinical outcomes post-treatment with EMR-based devices. One potential target is sirtuins, a novel family of genes thought to regulate life span and improve stress resistance by controlling key cellular metabolic and signaling pathways  and . Although long-term global reduced sirtuin expression has been shown to accelerate skin aging, its short-term potential role in mediating the local anti-aging effects of EMR-based device treatments remains largely unexplored. The goal of our study was to better understand the effects of FRF treatment on the sirtuin anti-senescence pathway in skin. Our results raise the possibility that sirtuin activity may be an additional quantitative corollary for treatment efficacy.
MATERIALS AND METHODS
This clinical study was conducted using a protocol approved by an institutional review board. Twenty-two healthy subjects who were 18 years or older and undergoing elective surgical face lift or abdominoplasty were enrolled in this study. Exclusion criteria included history of injection with fat, collagen, or synthetic material, bleeding disorder, prior, current, or anticipated treatment with anti-coagulants, thrombolytics, chemotherapeutics, systemic corticosteroids, anabolic steroids, or patients suffering from compromised immune system, impaired wound healing (e.g., diabetics and smokers), collagen vascular disease, and active infection.
Human skin tissue samples were obtained from voluntary subjects with prior informed consent within 40 minutes after their scheduled abdominoplasty procedure. Renesis™, a minimally invasive radiofrequency device developed by Primaeva Medical, Inc., was used to treat patients immediately, 2, 14, or 28 d prior to the abdominoplasty and biopsy to capture the temporal evolution of the in vivo wound healing response.
Preceding each treatment, skin was cleansed using 70% isopropyl alcohol, followed by wiping with topical 10% povidone iodine anti-septic. Subjects were then infiltrated with 1–2% lidocaine with or without 1:100,000 epinephrine. The FRF system was used to deliver bipolar RF energy to the dermis via 5 microneedle 30 gauge electrode pairs 6 mm in length, each spaced 1.25 mm apart. The angle of microneedle skin insertion was 20°. During RF energy application, the dermal tissue temperature within the treatment target zone was maintained at 72°C for 4 seconds using an intelligent feedback system. Superficial cooling to minimize epidermal damage was achieved using a solid-state Peltier device equipped with a heat sink and fan maintained at 15°C.
Semi-Quantitative RT-PCR Analysis
For RNA isolation from human skin tissue, frozen punch biopsies were thawed at room temperature, then homogenized in 2 ml of Trizol (Invitrogen, Carlsbard, CA) and mixed with 400 µl chloroform. After centrifugation, total RNA was extracted from the aqueous phase using the RNAeasy mini kit (Qiagen, Valencia, CA) following the manufacturer's protocol. Two micrograms of total RNA were used to synthesize cDNA using TaqMan Reverse Transcription Reagents (Applied Biosystems, Foster City, CA) and OligodT as primers. The total reaction volume was 50 μl. For RT-PCR, the annealing temperature was 24°C for 10 min, followed by first strand synthesis at 48°C for 1 h and heat inactivation at 95°C for 5 min. The SIRT1 primer was designed using Primer3 (WhiteHead Institute, Cambridge, MA) while all the remaining primers were obtained from the literature. The semi-quantitative RT-PCR reactions were performed on a DNA Engine Peltier Thermo Cycler (Bio-Rad, Hercules, CA). Reactions were carried out by activating DNA polymerase at 94°C for 2 min. The resultant cDNA was stored at −20°C. PCR was carried under the following conditions: denaturation at 94°C for 2 min and primer extension at 54°C for 30 s in 34 cycles for SIRT1-7, BAX, P53, KU70, FOXO3, β-ACTIN, and GAPDH.
Samples were electrophoresed on a 1.5% agarose gel containing 0.5 mg/ml of ethidium bromide and imaged using the Fluor Chem® HD2 Imaging System (Alpha Innotech Corporation, San Leandro, CA). Densitometry analysis was carried out using the AlphaEase FC software (Alpha Innotech Corporation). Intensity ratios were calculated as the intensity value for each gene divided by the intensity value of the internal control gene.
The means and standard errors of the amplicon intensity ratios for each gene of interest from a minimum of 3 independent runs were calculated using Microsoft Excel and statistical significance was determined using a paired analysis of variance. P values were taken to be statistically significant at P < 0.05.
Patients were treated immediately, 2, 14, or 28 d prior to their pre-scheduled abdominoplasty in order to capture the temporal evolution of the in vivo wound healing response. Biopsies for all time points were taken within 40 min post-abdominoplasty.
Determination of a Temporally Stable Internal Control
Gene of interest expression changes were compared against a temporally stable internal control, such that any variation in gene expression post-normalization would likely be due to the FRF-treatment rather than variations between samples. GAPDH and β-actin were both considered for this control. β-actin expression initially remained at baseline, but changed to 69%, 80%, and 192%; 2, 14, and 28 d post-FRF treatment, respectively. Conversely, GAPDH expression initially remained at baseline, rose to 118%, dropped to 97%, and rose to 117% 2, 14, and 28 d post-FRF treatment, respectively. GAPDH expression was shown to be relatively stable, not varying more than 18% from baseline with a max/min ratio of 1.2. β-actin expression was relatively unstable, varying up to 92% from baseline with a max/min ratio of 2.8. Thus, GAPDH was selected as the internal control.
Response of Sirtuin Genes to FRF Treatment
All the sirtuin genes except SIRT2 and 4 responded to FRF treatment (Fig. 1). SIRT1, 3, 5, and 7 were all downregulated immediately after FRF treatment to 56%, 59%, and 48%, of baseline, respectively. While SIRT3 remained roughly steady and below baseline for the rest of the time points, SIRT1 and 7 expression increased by more than twofold to 122% and 104% of baseline, respectively. SIRT1 remained roughly at the same level throughout the rest of the temporal series but SIRT7 showed a 40% increase to 148% by day 28. SIRT6, while remaining at baseline during the immediate phase, experienced a 45% increase to 153% by day 2, and increased to more than 200% of baseline by day 28.
Response of Apoptosis Genes to FRF Treatment
FoxO3, p53, and Bax responded to FRF treatment (Fig. 2). FoxO3 and p53 were downregulated immediately post-FRF treatment to 49.1% and 69.9% of baseline, respectively. From there, p53 expression steadily increased over the time series, showing a 50% increase between day 2 and 14, and ending at 138% of baseline by day 28. FoxO3 increased more than twofold by day 2 to 100%, remained steady through day 14, and then increased more than twofold again to 234% by day 28. Bax expression remained near baseline through day 14, but experienced a more than twofold increase to 204% by day 28. Ku70 showed no response to FRF treatment.
Response of MnSOD to FRF Treatment
MnSOD was immediately upregulated to 162% after treatment. By day 2, expression was downregulated back to baseline and remained there through day 28 (Table 1).
|Genes||Control||Day 0||Day 2||Day 14||Day 28|
The wound healing response has been established to follow a pattern of an initially strong inflammatory response and degradation of collagen followed by a gradual decrease in inflammation, apoptosis of immune cells, and collagen deposition . Sirtuins are a family of NAD+ dependent protein deacetylases that has been implicated in anti-aging and genetic stability pathways with specific links to genes and proteins such as Bax, Fork-head transcription factors, p53, and NF-κB, all of which are involved in the wound healing response [13-17].
Initial Proliferation, Inflammation, and Collagen Degradation
The immediate response to skin injury involves an immune response, wherein the injured area undergoes inflammation, and immune cells (such as macrophages) and ECM cells (such as fibroblasts) are recruited to the wound . FRF treatment results in an immediate drop in SIRT1 and 6 (Fig. 1), both of which inhibit the regulatory protein NF-κB . Increasing NF-κB allows for proliferation of the recruited immune and ECM cells, as it is closely involved in anti-apoptotic activity via reducing the activity of c-Jun N-terminal kinase . Similarly, the observed drop in p53 expression (Fig. 1) reflects an inhibition of apoptotic activity in the days immediately post-FRF treatment .
SIRT1, 6, and 7 levels also indicate a gradual increase in local inflammation in the days post-injury. The deregulation of NF-κB leads to an increase in inflammation . While the role of SIRT7 in the skin requires further elucidation, SIRT7 deficient mice have been shown to be less stress resistant and prone to inflammatory cardiomyopathy, suggesting the possibility of similar anti-inflammatory effects in skin . The observed downregulation of SIRT7 post-FRF treatment indicates that the protein no longer exerts its anti-inflammatory influence. Similarly, the initial decrease in FoxO3 expression can both be linked to the decreased expression of SIRT1, which induces FoxO3, and the increase in an inflammatory response, which results from decreased FoxO3 [24, 25] (Fig. 2).
Lastly, as SIRT1 has been shown to negatively regulate matrix metallopeptidases 1, 3, and 9, the decrease in SIRT1 expression immediately post-FRF treatment may have contributed to the observed increase in MMP production as well as the degradation of the old matrix and the remodeling of the dermal and subcutaneous collagen [8, 26, 27]. Altogether, the initial sirtuin expression noted in this experiment supports the hypothesis that the wound healing and immune response is heavily recruited immediately after FRF treatment.
Apoptosis, Collagen Deposition, Fibroblast Longevity
While the immediate response to FRF treatment is to induce an immune response, the effectiveness of this treatment, or indeed most treatments, is closely linked to the skin's ability to return the affected skin to normal function. A healthy wound healing response involves the restoration of cell populations in the area to normal levels through apoptosis of the recruited inflammatory immune cells, and inhibition of inflammation to prevent a runaway inflammatory response . The increase in SIRT1 and 6 expression in days 2–28 corresponds to the gradually increasing expression of FoxO3 and gradual inhibition of NF-κB, thereby promoting apoptotic activity over time while also inhibiting a runaway inflammatory response (Fig. 1) . Such increase in apoptotic activity is also reflected in the increase of p53 from day 2 onward (Fig. 2) . p53 also promotes SIRT6 expression, furthering its contribution to this effect .
Collagen production and protection of fibroblasts are also stimulated/induced by FRF treatment. Excess MMP production has been linked to skin aging and degradation of skin quality, thus suggesting that SIRT1 upregulation from days 2–28 post-FRF treatment acts as a protective measure to prevent the negative effects of superfluous MMP . Decreases in dermal fibroblast viability resulting in decreased collagen production and inhibition of procollagen biosynthesis neocollagenesis contribute significantly to poor skin quality and RF skin aging [30-32]. SIRT6 is capable of attenuating these negative effects through protecting dermal fibroblast genomic stability, and promoting collagen production in dermal fibroblasts [33, 34]. FRF resulting in neocollagenesis, as well as successful dermal remodeling in RF thermal zones, have been shown in previous studies by Berube et al.  and Hantash et al. . Altogether, it is evident that SIRT1 and 6 strongly contribute to anti-aging effects through regulation of MMPs and the promotion of fibroblast vitality.
Collagen denaturation and deposition is a necessary aspect of clinical improvement in skin quality, but is insufficient as a sole and direct corollary for effective treatment . Indeed, denaturation and deposition are steps involved in an overall comprehensive wound healing response, but alone are independent of aspects such as inflammation, proliferation or apoptosis of immune cells, and fibroblast longevity. Accordingly, a patient with a sustained reaction resulting in elevated levels of MMPs and fibroblast genomic instability may not see clinical improvement in skin quality, even with positive indications from looking solely at collagen levels. On the other hand, establishing sirtuin thresholds corresponding to a well regulated wound healing response may better correlate with improvements in skin quality, since sirtuin levels are intimately linked to not only collagen deposition and denaturation, but also to the immune response and fibroblast longevity. Future experiments will be focused on better understanding the relationship between sirtuin expression levels and clinical efficacy, allowing for appropriate treatment parameter selection. Insights from FRF may then be applied to other EMR modalities (ablative CO2, IPL, etc.). Clinically, this calibration may lead to a higher percentage of successful cosmetic treatments.